Diagnosis of timed automata with an application to industrial actuators

This paper concerns the modelling and diagnosis of discrete-event systems. The timed automaton is used as a discrete-event representation of such systems. The diagnosis is based on the investigation whether the measured input and output sequences are consistent with the timed automaton. The paper describes how the timed automaton can be obtained by qualitative identification using a set of measurement data and how it can be used together with the diagnostic algorithm to detect and identify faults. An important aspect of this method is the decomposition of the diagnostic system into the model and diagnostic algorithm which makes it possible to use the same algorithm for different systems by simply exchanging the model. The results are applied to industrial actuators.